Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/wdr.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     determine_ext,
   9     ExtractorError,
  10     js_to_json,
  11     strip_jsonp,
  12     unified_strdate,
  13     update_url_query,
  14     urlhandle_detect_ext,
  15 )
  16
  17
  18 class WDRBaseIE(InfoExtractor):
  19     def _extract_wdr_video(self, webpage, display_id):
  20         # for wdr.de the data-extension is in a tag with the class "mediaLink"
  21         # for wdr.de radio players, in a tag with the class "wdrrPlayerPlayBtn"
  22         # for wdrmaus, in a tag with the class "videoButton" (previously a link
  23         # to the page in a multiline "videoLink"-tag)
  24         json_metadata = self._html_search_regex(
  25             r'''(?sx)class=
  26                     (?:
  27                         (["\'])(?:mediaLink|wdrrPlayerPlayBtn|videoButton)\b.*?\1[^>]+|
  28                         (["\'])videoLink\b.*?\2[\s]*>\n[^\n]*
  29                     )data-extension=(["\'])(?P<data>(?:(?!\3).)+)\3
  30             ''',
  31             webpage, 'media link', default=None, group='data')
  32
  33         if not json_metadata:
  34             return
  35
  36         media_link_obj = self._parse_json(json_metadata, display_id,
  37                                           transform_source=js_to_json)
  38         jsonp_url = media_link_obj['mediaObj']['url']
  39
  40         metadata = self._download_json(
  41             jsonp_url, display_id, transform_source=strip_jsonp)
  42
  43         metadata_tracker_data = metadata['trackerData']
  44         metadata_media_resource = metadata['mediaResource']
  45
  46         formats = []
  47
  48         # check if the metadata contains a direct URL to a file
  49         for kind, media_resource in metadata_media_resource.items():
  50             if kind not in ('dflt', 'alt'):
  51                 continue
  52
  53             for tag_name, medium_url in media_resource.items():
  54                 if tag_name not in ('videoURL', 'audioURL'):
  55                     continue
  56
  57                 ext = determine_ext(medium_url)
  58                 if ext == 'm3u8':
  59                     formats.extend(self._extract_m3u8_formats(
  60                         medium_url, display_id, 'mp4', 'm3u8_native',
  61                         m3u8_id='hls'))
  62                 elif ext == 'f4m':
  63                     manifest_url = update_url_query(
  64                         medium_url, {'hdcore': '3.2.0', 'plugin': 'aasp-3.2.0.77.18'})
  65                     formats.extend(self._extract_f4m_formats(
  66                         manifest_url, display_id, f4m_id='hds', fatal=False))
  67                 elif ext == 'smil':
  68                     formats.extend(self._extract_smil_formats(
  69                         medium_url, 'stream', fatal=False))
  70                 else:
  71                     a_format = {
  72                         'url': medium_url
  73                     }
  74                     if ext == 'unknown_video':
  75                         urlh = self._request_webpage(
  76                             medium_url, display_id, note='Determining extension')
  77                         ext = urlhandle_detect_ext(urlh)
  78                         a_format['ext'] = ext
  79                     formats.append(a_format)
  80
  81         self._sort_formats(formats)
  82
  83         subtitles = {}
  84         caption_url = metadata_media_resource.get('captionURL')
  85         if caption_url:
  86             subtitles['de'] = [{
  87                 'url': caption_url,
  88                 'ext': 'ttml',
  89             }]
  90
  91         title = metadata_tracker_data['trackerClipTitle']
  92
  93         return {
  94             'id': metadata_tracker_data.get('trackerClipId', display_id),
  95             'display_id': display_id,
  96             'title': title,
  97             'alt_title': metadata_tracker_data.get('trackerClipSubcategory'),
  98             'formats': formats,
  99             'subtitles': subtitles,
 100             'upload_date': unified_strdate(metadata_tracker_data.get('trackerClipAirTime')),
 101         }
 102
 103
 104 class WDRIE(WDRBaseIE):
 105     _CURRENT_MAUS_URL = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/){1,2}[^/?#]+\.php5'
 106     _PAGE_REGEX = r'/(?:mediathek/)?[^/]+/(?P<type>[^/]+)/(?P<display_id>.+)\.html'
 107     _VALID_URL = r'(?P<page_url>https?://(?:www\d\.)?wdr\d?\.de)' + _PAGE_REGEX + '|' + _CURRENT_MAUS_URL
 108
 109     _TESTS = [
 110         {
 111             'url': 'http://www1.wdr.de/mediathek/video/sendungen/doku-am-freitag/video-geheimnis-aachener-dom-100.html',
 112             # HDS download, MD5 is unstable
 113             'info_dict': {
 114                 'id': 'mdb-1058683',
 115                 'ext': 'flv',
 116                 'display_id': 'doku-am-freitag/video-geheimnis-aachener-dom-100',
 117                 'title': 'Geheimnis Aachener Dom',
 118                 'alt_title': 'Doku am Freitag',
 119                 'upload_date': '20160304',
 120                 'description': 'md5:87be8ff14d8dfd7a7ee46f0299b52318',
 121                 'is_live': False,
 122                 'subtitles': {'de': [{
 123                     'url': 'http://ondemand-ww.wdr.de/medp/fsk0/105/1058683/1058683_12220974.xml',
 124                     'ext': 'ttml',
 125                 }]},
 126             },
 127         },
 128         {
 129             'url': 'http://www1.wdr.de/mediathek/audio/wdr3/wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100.html',
 130             'md5': 'f4c1f96d01cf285240f53ea4309663d8',
 131             'info_dict': {
 132                 'id': 'mdb-1072000',
 133                 'ext': 'mp3',
 134                 'display_id': 'wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100',
 135                 'title': 'Schriftstellerin Juli Zeh',
 136                 'alt_title': 'WDR 3 Gespräch am Samstag',
 137                 'upload_date': '20160312',
 138                 'description': 'md5:e127d320bc2b1f149be697ce044a3dd7',
 139                 'is_live': False,
 140                 'subtitles': {}
 141             },
 142         },
 143         {
 144             'url': 'http://www1.wdr.de/mediathek/video/live/index.html',
 145             'info_dict': {
 146                 'id': 'mdb-103364',
 147                 'ext': 'mp4',
 148                 'display_id': 'index',
 149                 'title': r're:^WDR Fernsehen im Livestream [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 150                 'alt_title': 'WDR Fernsehen Live',
 151                 'upload_date': None,
 152                 'description': 'md5:ae2ff888510623bf8d4b115f95a9b7c9',
 153                 'is_live': True,
 154                 'subtitles': {}
 155             },
 156             'params': {
 157                 'skip_download': True,  # m3u8 download
 158             },
 159         },
 160         {
 161             'url': 'http://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html',
 162             'playlist_mincount': 8,
 163             'info_dict': {
 164                 'id': 'aktuelle-stunde/aktuelle-stunde-120',
 165             },
 166         },
 167         {
 168             'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
 169             'info_dict': {
 170                 'id': 'mdb-1323501',
 171                 'ext': 'mp4',
 172                 'upload_date': 're:^[0-9]{8}$',
 173                 'title': 're:^Die Sendung mit der Maus vom [0-9.]{10}$',
 174                 'description': 'Die Seite mit der Maus -',
 175             },
 176             'skip': 'The id changes from week to week because of the new episode'
 177         },
 178         {
 179             'url': 'http://www.wdrmaus.de/filme/sachgeschichten/achterbahn.php5',
 180             'md5': '803138901f6368ee497b4d195bb164f2',
 181             'info_dict': {
 182                 'id': 'mdb-186083',
 183                 'ext': 'mp4',
 184                 'upload_date': '20130919',
 185                 'title': 'Sachgeschichte - Achterbahn ',
 186                 'description': 'Die Seite mit der Maus -',
 187             },
 188         },
 189         {
 190             'url': 'http://www1.wdr.de/radio/player/radioplayer116~_layout-popupVersion.html',
 191             # Live stream, MD5 unstable
 192             'info_dict': {
 193                 'id': 'mdb-869971',
 194                 'ext': 'flv',
 195                 'title': 'COSMO Livestream',
 196                 'description': 'md5:2309992a6716c347891c045be50992e4',
 197                 'upload_date': '20160101',
 198             },
 199         }
 200     ]
 201
 202     def _real_extract(self, url):
 203         mobj = re.match(self._VALID_URL, url)
 204         url_type = mobj.group('type')
 205         page_url = mobj.group('page_url')
 206         display_id = mobj.group('display_id')
 207         webpage = self._download_webpage(url, display_id)
 208
 209         info_dict = self._extract_wdr_video(webpage, display_id)
 210
 211         if not info_dict:
 212             entries = [
 213                 self.url_result(page_url + href[0], 'WDR')
 214                 for href in re.findall(
 215                     r'<a href="(%s)"[^>]+data-extension=' % self._PAGE_REGEX,
 216                     webpage)
 217             ]
 218
 219             if entries:  # Playlist page
 220                 return self.playlist_result(entries, playlist_id=display_id)
 221
 222             raise ExtractorError('No downloadable streams found', expected=True)
 223
 224         is_live = url_type == 'live'
 225
 226         if is_live:
 227             info_dict.update({
 228                 'title': self._live_title(info_dict['title']),
 229                 'upload_date': None,
 230             })
 231         elif 'upload_date' not in info_dict:
 232             info_dict['upload_date'] = unified_strdate(self._html_search_meta('DC.Date', webpage, 'upload date'))
 233
 234         info_dict.update({
 235             'description': self._html_search_meta('Description', webpage),
 236             'is_live': is_live,
 237         })
 238
 239         return info_dict
 240
 241
 242 class WDRMobileIE(InfoExtractor):
 243     _VALID_URL = r'''(?x)
 244         https?://mobile-ondemand\.wdr\.de/
 245         .*?/fsk(?P<age_limit>[0-9]+)
 246         /[0-9]+/[0-9]+/
 247         (?P<id>[0-9]+)_(?P<title>[0-9]+)'''
 248     IE_NAME = 'wdr:mobile'
 249     _TEST = {
 250         'url': 'http://mobile-ondemand.wdr.de/CMS2010/mdb/ondemand/weltweit/fsk0/42/421735/421735_4283021.mp4',
 251         'info_dict': {
 252             'title': '4283021',
 253             'id': '421735',
 254             'ext': 'mp4',
 255             'age_limit': 0,
 256         },
 257         'skip': 'Problems with loading data.'
 258     }
 259
 260     def _real_extract(self, url):
 261         mobj = re.match(self._VALID_URL, url)
 262         return {
 263             'id': mobj.group('id'),
 264             'title': mobj.group('title'),
 265             'age_limit': int(mobj.group('age_limit')),
 266             'url': url,
 267             'http_headers': {
 268                 'User-Agent': 'mobile',
 269             },
 270         }